// LLP-Antichain: maximum antichain by advancing chain indices to dominate-free positions.

class LLPAntichain {
  int[] LLPAntichain(int[][] chains, int[] len, boolean[][] leq) {
    int[] G = 0;
    forbidden (j) :
      G[j] < len[j] && (exists k in [0..n-1] : k != j && leq[chains[j][G[j]]][chains[k][G[k]]])
    =>
      advance :
        G[j] = G[j] + 1;
    return G;
  }
}